-
Notifications
You must be signed in to change notification settings - Fork 0
/
4_process_energy_costs.R
299 lines (184 loc) · 10.6 KB
/
4_process_energy_costs.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
# 4.1 import the PV cost layers
rr <- list.files(paste0(input_folder, "pv_cost"), full.names = T, pattern = "tif")
df <- data.frame(rr)
df$name <- gsub(".tif", "", df$rr)
df$name <- sub('.*/', '', df$name)
df$crop <- sub("\\&.*", "", df$name)
df$irr <- sub('.*&', '', df$name)
# 4.2 cluster the MapSPAM crops to match the crop-technology couplets from the Earth's future paper to all the crops
crops = readxl::read_xlsx(paste0("H:/ECIP/Falchetta/MLED_database/" , 'crops_cfs_ndays_months.xlsx')) %>% dplyr::select(crop, eta_irr)
crops$most_sim <- c("Wheat", "Banana", "Maize", "Banana", "Maize", "Maize", "Onions", "Wheat", "Maize", "Commonbeans", "Sugarcane", "Wheat", "Onions")
crops$irr_type <- ifelse(crops$eta_irr==0.6, "flood", "drip")
crops <- merge(df, crops, by.x=c("crop"), by.y=c("most_sim"))
# prevalent crop in each grid cell
files = list.files(path = paste0(input_folder, "spam_folder/spam2017v2r1_ssa_harv_area.geotiff") , pattern = 'R.tif', full.names = T)
files <- files[tolower(as.character(substr(basename(files), 20, 23))) %in% crops$crop.y]
for (X in files){
a = paste0("A_" , tolower(as.character(substr(basename(files), 20, 23))))
clusters[a] <- exactextractr::exact_extract(raster(X), clusters, fun="sum")
}
aa <- clusters
aa= st_set_geometry(aa, NULL)
clusters$maxq <- colnames(aa)[grepl("A_", colnames(aa))][apply(aa[grepl("A_", colnames(aa))],1,which.max)]
clusters$maxq <- gsub("A_", "", clusters$maxq)
clusters$maxq <- gsub("_r", "", clusters$maxq)
clusters <- merge(clusters, data.frame(crops[,c(2, 5)]), by.y="crop.y", by.x="maxq", all.x=T)
clusters <- clusters[!duplicated(clusters$cl_id), ]
clusters$rr <- ifelse(is.na(clusters$rr), names(sort(-table(clusters$rr)))[1], clusters$rr)
##
list <- as.character(unique(clusters$rr))
list <- ifelse(is.na(list), "H:/ECIP/Falchetta/MLED_database/input_folder/pv_cost/Maize&flood.tif", list)
list2 <- pblapply(list, raster)
clusters$pvcost <- NA
for (i in 1:length(list2)){
clusters[names(list2[[i]])] <- exact_extract(list2[[i]], clusters, fun="mean")
}
for (i in 1:nrow(clusters)){
clusters$pvcost[i] <- pull(clusters[i,(which(grepl("pvcost", colnames(clusters))))+match(as.character(clusters$rr[i]), list)] %>% st_set_geometry(NULL))
}
# extrapolate where NA
# extract solar radiation
files <- list.files(path=paste0(input_folder, "pv_cost"), pattern="asc", full.names = T)
for (i in 1:length(files)){
sr <- raster(files[i])
clusters[paste0("sr_", i)] <- exact_extract(sr, clusters, "mean")
}
# regress on solar radiation and prevalent crop and X and Y and predict
clusters <- bind_cols(clusters, data.frame(st_coordinates(st_centroid(clusters))))
model <- lm(pvcost ~ sr_1 + sr_2 + sr_3 + sr_4 + sr_5 + sr_6 + sr_7 + sr_8 + sr_9 + sr_10 + sr_11 + sr_12 + X*Y , data=clusters)
#summary(model)
pred <- predict(model, clusters)
clusters$pvcost <- ifelse(is.nan(clusters$pvcost), pred, clusters$pvcost)
###
if(nrow(clusters)!=nrow(read.csv("pvout_t.csv"))){pvout_t <- stack(pblapply(list.files(paste0(input_folder, "monthly"), full.names = T, pattern = "asc"), raster)) ; pvout_t <- exact_extract(pvout_t, clusters, "mean"); pvout_t <- (pvout_t/30) * 0.65; write.csv(pvout_t, "pvout_t.csv")} else{
pvout_t <- read.csv("pvout_t.csv")
pvout_t$X<- NULL
}
##
clusters <- merge(clusters, vat_import, by.x="sov_a3", by.y="ISO3", all.x=T)
clusters$VAT_PV <- ifelse(is.na(clusters$VAT_PV), mean(vat_import$VAT_PV), clusters$VAT_PV)
clusters$VAT_BATTERY <- ifelse(is.na(clusters$VAT_BATTERY), mean(vat_import$VAT_BATTERY), clusters$VAT_BATTERY)
clusters$IMPORT_PV <- ifelse(is.na(clusters$IMPORT_PV), mean(vat_import$IMPORT_PV), clusters$IMPORT_PV)
clusters$IMPORT_BATTERY <- ifelse(is.na(clusters$IMPORT_BATTERY), mean(vat_import$IMPORT_BATTERY), clusters$IMPORT_BATTERY)
clusters$totalcost = NA
clusters$landuse_m2 = NA
clusters$pvsize_kw = NA
clusters$batterysize_kwh = NA
clusters$share_pvout_used_for_pumping = NA
clusters$residual_pvout_notused_for_pumping = NA
library(insol)
dem_t <- clusters %>% dplyr::select(starts_with("er_kwh")) %>% st_set_geometry(NULL)
dem_t <- dem_t[,c(1:12)]
set_obs <- which(clusters$er_kwh_tt>0)
for (ncls in set_obs){
lat = clusters$Y[ncls]
long = clusters$X[ncls]
powerpump <- clusters$powerforpump[ncls] # power of the pump, in KW
dem <- as.numeric(dem_t[ncls,]) / (30/irrigation_frequency_days) # daily demand, in kwh for each day of each month
pvout <- as.numeric(pvout_t[ncls,]) # daily generation potential (kWh/kWp) for each day of each month
sunlighthours <- (daylength(lat=lat,long=long,JD(seq(ISOdate(2019,1,1),ISOdate(2019,12,31),by='month')),tmz=1)[,3]) * 0.65 # number of sunlight hours at 100% potential equivalent ("peak sun") per month
usehours_which <- which(load_curve_irrig!=0) # number of hours when the pump is operating
usehours <- length(usehours_which)
mat <- daylength(lat=lat,long=long,JD(seq(ISOdate(2019,1,1),ISOdate(2019,12,31),by='month')),tmz=1)
out <- matrix(ncol=length(usehours_which), nrow=12)
k = 1
for (i in usehours_which){
out[,k] <- cbind( mat[,1] <= i & mat[,2] >= i )
k = k+1
}
if (scenarios$water_tank_storage[scenario]==T){
out <- ifelse(out==FALSE, TRUE, TRUE)
}
demand_no_supply_hours <- usehours - rowSums(out) # number of hours when the pump use is not in sunlight hours
######################
if (scenarios$water_tank_storage[scenario]==T){
toprovided_withbatteries_daily <- 0.2 * dem # demand to be supplied when no generation potential is available
} else{ toprovided_withbatteries_daily <- (demand_no_supply_hours/usehours) * dem
}
########
if (scenarios$no_battery[scenario]==F){
toprovided_withbatteries_daily <- 0
} else{ toprovided_withbatteries_daily <- (demand_no_supply_hours/usehours) * dem
}
if (scenarios$no_battery[scenario]==F){
toprovided_directly_from_pv_daily <- dem # demand to be supplied directly via PV
} else{ toprovided_directly_from_pv_daily <- (1-(demand_no_supply_hours/usehours)) * dem
}
# pv size
kwh_per_hour_direct_use <- toprovided_directly_from_pv_daily / (usehours - demand_no_supply_hours)
wp_direct_use <- (kwh_per_hour_direct_use / (pvout/sunlighthours)) * 1000 * 1.1 # energy lost in the system
kwh_per_hour_battery <- toprovided_withbatteries_daily / (sunlighthours - (usehours - demand_no_supply_hours))
wp_to_charge_battery <- (kwh_per_hour_battery / (pvout/(sunlighthours - (usehours - demand_no_supply_hours)))) * 1000 * 1.2 # energy lost in the system
wp <- max(wp_direct_use, wp_to_charge_battery) # required pv size
kwp <- wp/1000
# actual use of power for pumping
share_pvout_used_for_pumping <- sum((kwh_per_hour_direct_use * usehours), na.rm=T) / sum(((pvout * kwp) - (kwh_per_hour_direct_use * usehours)), na.rm=T)
clusters$share_pvout_used_for_pumping[ncls] = ifelse(!is.finite(share_pvout_used_for_pumping), NA, share_pvout_used_for_pumping)
residual_pvout_notused_for_pumping <- (pvout * kwp) - (kwh_per_hour_direct_use * usehours)
clusters$residual_pvout_notused_for_pumping[ncls] = sum(residual_pvout_notused_for_pumping * 30, na.rm=T)
# battery size
kwh_battery <- max((toprovided_withbatteries_daily / 0.85 / 0.8)) # required battery size, net of battery efficiency, depth of discharge, extra size for unexpected low generation buffer
kwh_battery <- kwh_battery + (max(dem)*1) # extra days of autonomy
# inverter size
inverter_size <- powerpump * 1.3 #W
## plots
# plot(dem, type="l", col="red")
# plot(pvout, type="l", col="blue")
# plot(sunlighthours, type="l", col="red")
# plot(toprovided_directly_from_pv_daily, type="l", col="red")
# plot(toprovided_withbatteries_daily, type="l", col="red")
# plot(kwp, kwh_battery)
################
# 2. estimate costs of installing the system and O&M
# solar + battery costs
usdperkwhbattery <- 500
usdperkwhbattery_lower_10 <- 500 * 0.9
usdperkwhbattery_lower_25 <- 500 * 0.75
clusters$pvcost_lower_10 <- clusters$pvcost * 0.9
clusters$pvcost_lower_25 <- clusters$pvcost * 0.75
if (scenarios$pvbatterycosts_sens[scenario]=="baseline"){
usdperkwhbattery <- usdperkwhbattery
clusters$pvcost <- clusters$pvcost
} else if(scenarios$pvbatterycosts_sens[scenario]=="minus10pct"){
usdperkwhbattery <- usdperkwhbattery_lower_10
clusters$pvcost <- clusters$pvcost_lower_10
} else{
usdperkwhbattery <- usdperkwhbattery_lower_25
clusters$pvcost <- clusters$pvcost_lower_25
}
##########
##########
if (scenarios$instalments_business_model[scenario]==4 | scenarios$instalments_business_model[scenario]==3){
pvcost <- npv(discount_rate, rep(((wp * clusters$pvcost[ncls] ) / lifetimepump)), 0:(lifetimepump-1))
storcost <- npv(discount_rate, rep(((kwh_battery * usdperkwhbattery) / lifetimepump)), 0:(lifetimepump-1))
} else{
pvcost <- wp * clusters$pvcost[ncls]
storcost <- kwh_battery * usdperkwhbattery
}
if (scenarios$VAT_import_costs[scenario]==T){
pvcost <- pvcost * ((1 + clusters$VAT_PV[ncls]/100)) + (1*(clusters$IMPORT_PV[ncls]/100))
storcost <- storcost * ((1 + clusters$VAT_BATTERY[ncls]/100)) + (1 + (clusters$IMPORT_BATTERY[ncls]/100))
}
if (scenarios$no_battery[scenario]==F){
storcost <- 0
}
om_cost <- npv(discount_rate, rep((pvcost*0.1), lifetimepump), 0:(lifetimepump-1))
inverter_cost <- inverter_size * 0.2 #usd per watt
inverter_cost <- ifelse(clusters$gr_wat_depth[ncls]<=10 | clusters$which_pumping[ncls]=="Surface water pumping", 0, inverter_cost) #DC used if water is easily accessible, so no need for an inverter
ch_contr_cost <- (pvcost + om_cost + storcost + inverter_cost) * 0.1
totalcost <- (pvcost + om_cost + storcost + ch_contr_cost + inverter_cost) * clusters$npumps[ncls]
#
landuse_m2 <- (kwp * 5 + ifelse(kwp>0, 9, 0)) * clusters$npumps[ncls] # m2 + space for charge controller, inverter, pump
clusters$totalcost[ncls] = totalcost
clusters$landuse_m2[ncls] = landuse_m2
clusters$pvsize_kw[ncls] = kwp
clusters$batterysize_kwh[ncls] = kwh_battery
}
clusters$totalcost = ifelse(clusters$yearly_IRREQ==0, NA, clusters$totalcost)
clusters$pvsize_kw = ifelse(clusters$yearly_IRREQ==0, NA, clusters$pvsize_kw)
clusters$batterysize_kwh = ifelse(clusters$yearly_IRREQ==0, NA, clusters$batterysize_kwh)
clusters$landuse_m2 = ifelse(clusters$yearly_IRREQ==0, NA, clusters$landuse_m2)
summary(clusters$totalcost[clusters$totalcost>0]/clusters$npumps[clusters$totalcost>0], na.rm=T)
summary(clusters$pvsize_kw[clusters$pvsize_kw!=0])
summary(clusters$batterysize_kwh[clusters$batterysize_kwh!=0])
write_rds(clusters, paste0("clusters_with_data_4_", paste(scenarios[scenario,], collapse="_"), ".Rds"))